The overall goal of this project is to achieve a better understanding of the processes whereby intentions are translated into actions: "executive" processes. The specific aims of this project are to study the executive processes involved in 1) switching from one task to another, and 2) preparing such a switch. These are important aspects of executive processes inasmuch as normal humans switch between tasks frequently and easily whereas specific patient populations are unable to: their actions become "captured" by the stimuli in the environment. We propose to address this issue by studying the electrophysiological (ERP) correlates of the executive processes involved in task-switching in normal subjects, as well as the underlying functional architecture of the system(s) involved (using fMRI). We propose to study these issues in 4 experiments. In the first, we will investigate a new ERP component that we have discovered that appears to be an index of "task (response?) readiness" inasmuch as its amplitude and subjects' RTs appear to closely related. In the second experiment, we will investigate this issue further, using fMRI to localize the area(s) responsible for generating this component. In the third experiment, we will investigate cued switches of task, and the ERP componentry associated with preparing to switch. This will be compared to the componentry associated with actually performing a switch, as well as that associated with preparing to repeat the same task. In the final experiment, we will use fMRI to investigate the network underlying the ERP componentry. In pilot work, we have found an ERP component associated with the cue that occurs at ~350 ms, regardless of when the imperative stimulus is presented. This suggests that there are some task-switching-related processes that require a set period of time to be completed. If this is so, it would go some way towards explaining the decrease in "switch costs" as subjects are allowed more time to "prepare" for the forthcoming task. Thus, our pilot data suggest that we have found two robust methods for investigating control processes, and it is the purpose of this proposal to use these paradigms to gain a better understanding of executive control. [unreadable] [unreadable]